Predicting environmental suitability and geographical distribution of Dicrocoelium dendriticum at littoral of Caspian Sea: An ecological niche-based modeling

Dicrocoeliasisis caused by the small liver fluke (Dicrocoelium spp.), mainly Dicrocoelium dendriticum in domestic and wild ruminants. The small liver fluke is the probable predisposing cause of economic burden. The impact of geographic and climatic factors on the incidence of dicrocoeliasis has been severely ignored in different geographical areas. Due to the lack of data regarding dicrocoeliasis in Iran, this study was aimed to investigate the prevalence and intensity of ovine and bovine Dicrocoelium infection in the coastal strip south of the Caspian Sea. Fecal samples were obtained from the cattle and sheep in three provinces of Guilan, Mazandaran and Golestan at the littoral of the Caspian Sea. All collected samples were then tested by flotation methods for determining the number of eggs per gram of feces (EPG). Moreover, we applied maximum entropy niche-based modeling (MaxEnt), coupled with remote sensing and the Geographical Information System (GIS) to visualize the spatial distribution and risk factors of Dicrocoelium dendriticum at the littoral of Caspian Sea. A total of 2688 stool samples were collected from cattle (n = 1344) and sheep (n = 1344) in coastal provinces of the Caspian Sea including Guilan (n = 1280), Mazandaran (n = 768) and Golestan (n = 640) provinces. Based on the data presented here, the highest rate of infection was observed in Guilan and Mazandaran provinces. The results revealed the prevalence rates of 36.72% and 6.09% for sheep and cattle in Guilan province, respectively. This rate was 22.4% for sheep and 3.91% for cattle in Mazandaran province. However, the rate of sheep infection was 90% in some point locations. Dicrocoelium infection was found to be significantly different between three provinces in sheep (P < 0.00001, Chi = 111.633). Our findings exhibited a high reliability of the MaxEnt model, and area under the curve (AUC) values of the training and test data sets were determined to be 0.852 and 0.818, respectively. Jackknife analysis showed the relative variable contribution to the model performance, where four variables were found as key influential factors that highly affected the habitat suitability of the presence of the lancet fluke including the precipitation of driest quarter (Bio17), altitude, temperature seasonality (Bio4), and precipitation of driest month (Bio14). The findings of this study demonstrated a high presence rate of Dicrocoelium infection at the littoral of Caspian Sea, Iran. Moreover, climatic variables can be considered as important predictive factors affecting the distribution of infection in the studied areas. Further studies based on the findings of the GIS are also very important in the country for planning control programs.